Large Language Models Are State-of-the-Art Evaluator for Grammatical Error Correction

Autor:	Kobayashi, Masamune, Mita, Masato, Komachi, Mamoru
Rok vydání:	2024
Předmět:	Computer Science - Computation and Language
Druh dokumentu:	Working Paper
Popis:	Large Language Models (LLMs) have been reported to outperform existing automatic evaluation metrics in some tasks, such as text summarization and machine translation. However, there has been a lack of research on LLMs as evaluators in grammatical error correction (GEC). In this study, we investigate the performance of LLMs in GEC evaluation by employing prompts designed to incorporate various evaluation criteria inspired by previous research. Our extensive experimental results demonstrate that GPT-4 achieved Kendall's rank correlation of 0.662 with human judgments, surpassing all existing methods. Furthermore, in recent GEC evaluations, we have underscored the significance of the LLMs scale and particularly emphasized the importance of fluency among evaluation criteria. Comment: Accepted to BEA workshop at NAACL 2024
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2403.17540 Zobrazit plný text záznamu View this record from Arxiv